Abstracts

Rationale: Recurrent epileptic seizures involve the hippocampus in the brain. We hypothesized that pathological synchronous behavior of neurons occurs during the genesis of epilepsy. Further, we hypothesized that propagation of seizures takes place in both transverse and longitudinal directions in the hippocampus CA1. We tested above hypothesis by using quantitative methods to measure the synchrony and seizure propagation during electrically-kindled seizures. Methods: Adult male Sprague-Dawley rats (250-300g) were studied. A bipolar stimulation electrode was stereotactically implanted in hippocampal CA3 region. A four-channel microelectrode array was placed to record neuronal activities in contralateral hippocampal CA1 region along the transverse (lamellar) or longitudinal (septotemporal) hippocampal axis. One channel of each microelectrode array was filtered (0.1-5,00 Hz) to record raw EEG, whereas, in parallel, all the channels were filtered (300-4,000 Hz) and sampled at 25 kHz to record high frequency activates. Kindled seizures were initiated by using a standard stimulation protocol (10s, 50Hz, 1ms biphasic pulses), and behavioral seizures were scored according to Racine stages. Synchrony between simultaneously recorded activities was quantified by theta phase analyses and event synchronization (ES). Raw EEG signal was band-pass filtered between 4-10Hz to extract theta oscillation. Theta phase approaches work directly with the timing of neuronal discharges by assigning phase value to the waveform. Shannon entropy was used to quantify the order of theta phase distribution of the waveform during the seizures. ES quantifies the number of times a peak occurs in two recordings within a small window. Peaks that occur in both recordings within a time interval are considered to be synchronous. Results: We recorded 150 kindled seizures from 5 freely moving rats with transverse microelectrode array configuration. The mean Shannon entropy was 1.959±0.033 for stage 1 seizures, and lowered to 1.509±0.097 for stage 5 seizures, which showed neuronal synchrony increased as kindling progressed in CA1 lamellar direction. Theta phase analyses also demonstrated that CA1 pyramidal cells fired preferentially around the negative peak of theta oscillation with increasing behavioral seizure scores. Propagation pattern, revealed by calculating phase differences, showed that kindling from CA3 region resulted in neuronal discharges in CA1along lamellar from proximal to distal. Seizures (n=145) recorded from longitudinally placed electrodes showed no significant change in Shannon entropy as kindling progressed. Further, no preferred timing was observed between neuronal firing and theta oscillation. Finally, neuronal firing revealed no propagation in septotemporal direction. Event synchronization showed neuronal firings along lamellar direction were more correlated (ES strength= 0.67) compared to septotemporal direction (ES strength=0.36). Conclusions: During kindled seizures, neuronal firings in CA1 are synchronized and propagate along its lamellar axis, but not in its septotemporal direction.